Integration of EHR and Genomic Data for Personalized Cancer Treatment

Abstract

Introduction The integration of Electronic Health Records (EHR) and genomic profiling represents a transformative 
advancement in precision oncology, enabling biomarker-driven therapy selection, real-time treatment adaptation, and 
improved clinical outcomes. This study evaluated the impact of EHR–genomic integration on therapeutic decision-making, 
progression-free survival (PFS), and clinician workflow efficiency in a multi-cancer cohort. 
Participants: A total of 1,200 adult oncology patients with histologically confirmed malignancies, ECOG performance 
status 0–2, and availability of tumor tissue or plasma for sequencing were enrolled from January 2022 to December 2024. 
Cancer types included NSCLC, breast, colorectal, ovarian, melanoma, and other solid tumors, with 86% presenting at 
AJCC Stage III–IV. 
Instruments: Clinical data were extracted from HL7 FHIR–compliant EHR systems using SNOMED CT and LOINC 
terminologies, supplemented by NLP-driven parsing of unstructured clinical notes. Genomic profiling utilized FFPE or 
fresh frozen tumor tissue and plasma cfDNA, sequenced on Illumina NovaSeq platforms with targeted panels (TruSight 
Oncology 500, FoundationOne CDx). Bioinformatics pipelines incorporated BWA-MEM, GATK, CNVkit, STAR-Fusion, 
and variant annotation via ClinVar, COSMIC, and OncoKB. 
Procedure: Clinical and genomic datasets were harmonized in an OMOP Common Data Model repository under 
HIPAA/GDPR compliance. An EHR-embedded Clinical Decision Support System (CDSS) generated therapy 
recommendations mapped to NCCN/ESMO guidelines, reviewed by multidisciplinary tumor boards. Treatment 
adjustments were documented, and follow-up data were collected at defined intervals. 
Data Analysis: Kaplan–Meier survival analysis and Cox proportional hazards modeling assessed the effect of integrated 
care on PFS. Chi-square and t-tests compared categorical and continuous variables, respectively, with significance set at p 
< 0.05. Pre- and post-integration outcomes, including targeted therapy eligibility, time-to-treatment decision, response 
rates, and adverse drug reaction incidence, were compared. 
Interrater Interpretability: Two independent oncologists and one genetic counselor reviewed genomic report 
interpretations, with interrater agreement measured using Cohen’s kappa. Agreement exceeded κ = 0.85 for Tier I and II 
actionable variants, confirming high interpretive consistency across clinical reviewers...